Gym Retention Strategies - Study of Habit Formation
Mafalda Filipa Baptista de Oliveira
Master Thesis
Supervisor at FEUP: Prof. Maria Gabriela Beirão dos Santos Supervisor at Solinca Health & Club: Alexandre Freitas
“What makes the desert beautiful,' said the little prince, 'is that somewhere it hides a well…” Antoine de Saint-Exupéry's masterpiece The Little Prince
Abstract
There are several researches in the literature related to the study of habit formation and customer retention, but it wasn’t found any evidence of one that have explored those topics in a gym context. In addition, this study aims to analyze if adding specific rewards to customer subscription may have impact on habit formation; and if the more frequently the customer attends to the gym, the stronger the habit becomes.
More than half of the gym members cancel the contract in the first 12 months (in low-cost players can reach 90%), most of them for low use of the gym. Especially for new users, this is due to the non-creation of the training habit at the beginning of the life cycle and then only aspects such as price, competition or service-specific factors (dissatisfaction with facilities, cleaning, maintenance, etc.).
The increasing importance given to the health and wellness issues makes this topic very relevant and challenging, offering an opportunity for gyms to understand how habit formation can influence retention, so that they can come up with strategic action plans that leads to an increase of regular physical activity habits in the gym.
This research follows a quantitative methodology approach based on analysis of gym’s customer database (on one year period) and a paper-based questionnaires delivered to gym’s customers. The outcomes obtained from data analysis revealed a positive influence of rewards on habit formation and a positive association between habit formation and customer retention. In addition, it was also found an increasing strength of habit for higher gym frequency. These influences were expected and supported by the existing literature.
Acknowledgments
This space I dedicate to all those who, in a more or less direct way, contributed to the conclusion of this stage with the sense of accomplishment.
I would like to thank to my colleagues from Solinca & Health Club for all the dedication and companionship throughout the project, and for helping me to move forward in deadlock moments.
Moreover, because I could not fail to do so, I thank my supervisor, Prof. Maria Gabriela Beirão dos Santos, for her dedication and for being always available to advise me and accompany me throughout the process. Without her encouragement, I certainly would not have been able to complete yet another stage of my academic education with the quality that so requires.
Finally, I leave a deep gratitude to my family - especially my mother, my sisters and my grandmother - and also to my friends who supported me along this journey, and always “got my back” and believed in this happy ending.
Table of Contents
1 Introduction ... 1
1.1 Project background ... 1
1.2 Problem Description ... 1
1.3 Research Questions ... 2
1.4 Study and Project Development at Solinca Health & Club ... 3
1.5 Report outline ... 3
2 Literature Review ... 5
2.1. Definitions of Habit ... 5
2.2. Habit formation stages ... 7
2.3. Habit Formation Obstacles ... 8
2.4. Customer Retention ... 9
2.5. Literature review conclusion ... 11
3 Problem Characterization ... 12
4 Methodology ... 16
4.1 Comparative analysis of existing approaches and reasons for the choice of adopted approach... 16
4.2 Method used in the project ... 17
4.2.1 Database Analysis ... 17
4.2.1.1 Database Sample Characterization ... 18
4.2.2 Survey Instrument ... 19
4.2.2.1 Survey Sample Characterization ... 20
5 Results ... 21
5.1. Database results ... 21
5.2 Survey results ... 25
5.3 Research questions’ analysis ... 27
6 Conclusion and future research ... 30
References ... 34
List of Tables
Table 1 Existing definitions of habit ... 6
Table 2 Top 20 Worldwide Fitness Trends for 2018... 14
Table 3 Gym clients’ sample characterization ... 18
Table 4 New membership subscriptions in a one-year period ... 19
Table 5 Survey sample characterization ... 20
Table 6 Database sample characterization ... 22
Table 7 Association between gender and habit ... 23
Table 8 Association between group classes and habit formation ... 23
Table 9 Principal component analysis ... 25
Table 10 Sample's intention to attend more times the gym ... 27
Table 11 Crude and adjusted OR for incentive impact on habit ... 28
Table 12 Association between habit formation and retention ... 28
Table 13 Strength of habit formation comparison according to gym frequency ... 29
List of Figures
Figure 1 Portuguese's favourite gym activities - source Público newspaper, in 2016 ... 12 Figure 2 Fitness industry in Portugal (2016) ... 13 Figure 3 Individual and smooth spline lines for the number of entries along the 31 fortnights
... 24 Figure 4 Smooth splines for the number of entries along the 31 fortnights for gender... 24 Figure 5 Smooth splines for the number of entries along the 31 fortnights for age categories 24 Figure 6 Reasons that prevent gym attendance ... 26
List of abbreviations
1 Introduction
1.1 Project background
This project was developed in a business environment, in a company of the fitness industry (Solinca Health & Club), within the scope of the Master in Services Engineering and Management of the Faculty of Engineering of the University of Porto.
Currently, Solinca Health & Club is one of the largest fitness chains in Portugal with 20 active clubs and began its activity in 1995 with the opening of its first fitness club in Porto Palácio Hotel & SPA. In 1997, Solinca Health & Club opened another fitness unit in Colombo Shopping Center, which is currently the largest club in Greater Lisbon. In 2001 were inaugurated Solinca Vasco da Gama and Solinca Braga, in 2003 was time for Solinca Oeiras and, the following year, Viana do Castelo. It was followed by Solinca NorteShopping in Matosinhos in 2006 and Solinca Dragão in Porto, which is the largest Solinca Club in the country (with 6,800 m2). In 2011 it was Solinca Gaia's turn, and more recently, Solinca Vila do Conde, Solinca Alfragide, Solinca Maia, Solinca Guimarães, Solinca Foz, Solinca Ermesinde and Solinca Laranjeiras (by acquisition), Solinca Constituição, Solinca Loures, Solinca Lumiar and Solinca Rio Tinto.
Gyms face the problem of membership cancelations within the first 12 months, mostly due to low use of the gym. Some reasons are the non-creation of training habit and the lack of motivation. Much of our lives are ruled not by our conscious decisions or thoughts, but by our habits. Once embedded, the very stickiness of habits means they are strong and hard to displace. Trying to clear up existing habits and getting people to change and do something different can be a major mental battleground, trying to initiate a new habit, is something completely different.
So, having that in mind, Solinca Health & Club is well aware of the challenges that fitness faces and therefore the aim of this project is to study habit formation and customer retention in fitness industry, in order to come up with some marketing actions proposals to increase retention.
1.2 Problem Description
More than half of the gym members cancel the contract in the first 12 months (in low-cost players can reach 90%), most of them for low use of the gym. Especially for new users, this is due to the non-creation of the training habit at the beginning of the life cycle and the lack of
motivation inherent in not fulfilling the proposed objectives - and then only aspects such as price, competition or service-specific factors (dissatisfaction with facilities, cleaning, maintenance, etc.).
As previously mentioned, Solinca Health & Club is one of the largest fitness chains in Portugal with 20 active clubs (with approximately 60.000 active members), so it is natural that it has a very present and aware concern to the issue of retention in their fitness Clubs, and has the experience and empirical knowledge that much of the retention rates observed are due to the formation or the lack of it.
Like any other company, Solinca Health & Club is concerned with retention and churn rates and is trying to overcome these possible events through acknowledge the reasons why the individuals choose a specific fitness Club. Other reasons like customer’s motivation, expectations, evaluation of their behavior, diagnosis of the current lifestyle, prescription of the exercise according to the needs and tastes, are also in line in order to satisfy them better. The central issue of this project focuses on the fact that individuals do not always form the habit of attending the gym to practice exercise, and so end up by attending fewer and enter the so-called fitness cliff. To better address this situation, in addition to the initiatives outlined above, Solinca Health & Club decided to put into practice a marketing strategy (“Money Back”) that would help individuals to form the habit of attending the gym. Thus, all members who in the first 30 days of their contract made 12 accesses at the fitness Club, would see returned half of the membership fee they paid when they signed up. So, with some encouragement, Solinca Health & Club believes it can be a good start in the new routine of individuals in order to keep them healthy and with an active lifestyle.
1.3 Research Questions
This research aims to better understanding the habit concept and formation and its impact on customer retention in the fitness industry. More specifically aims to address the following research questions.
1. What is the impact of rewards on habit formation?
2. How habit formation in fitness industry influences customer retention? 3. What is the impact of gym frequency on habit strength?
1.4 Study and Project Development at Solinca Health & Club
This project began by learning the current business model of Solinca Health & Club, focusing specifically on the processes that involve the integration of the individual in the fitness Club, as well as the marketing strategies available in this process.
Then, the history information in Solinca Health & Club’s customer database was analyzed, from February 2017 to March 2018, in order to analyze people’s behavior in the different fitness Clubs, analyzing for instance the average attendance in the clubs, presence in group classes, subscription of personal training services, frequency type (schedule restrictions or not). This way, it will be possible to understand to what extent the mentioned factors serve as a measure for the habit formation in the individuals and thus to analyze their behavior regarding the retention in the club. The retention at Solinca Health & Club is analyzed according to the contract time, and for that reason, for the due effects it was considered a year of analysis in order to perceive if the individuals finishing this time, renew or not the service. On the other hand, it is also intended to study the formation of habit, in the sense that after being repeated several times, it can become automatic, that is, with little or no decision of the individual, and can still be part of he’s identity or personal style. For this purpose, a paper-based survey was distributed to clients at Solinca Health & Club Maia (where the project was developed), between 16th and 29th may 2018, to find out if gym frequency can strengthen the habit.
From the information obtained in the previous steps, and based on the review of the literature conducted some assumptions and suggestions were highlighted on the topic under study.
1.5 Report outline
The present dissertation consists of six chapters.
In Chapter 1 - Introduction, all relevant data will be provided for the general framework of the subject and understanding of the problem under study, as well as the structure of the dissertation.
Regarding Chapter 2 - Literature Review, this includes the theoretical foundation of all the concepts that are considered pertinent to the study, being: definitions of habit, Habit formation stages, habit formation obstacles and customer retention; as well as the main contributions of this dissertation.
As to Chapter 3 - Problem Characterization, this consists of the characterization of the problem, including research objectives.
In Chapter 4 - Methodology, describes the methodology used, with reference to the data collection instruments and the selection of the sample.
In Chapter 5 - Results, we present the results obtained from the methodology used.
Finally, Conclusion and future research are presented in Chapter 6, which includes discussion of results, study contributions, limitations and suggestions for future research.
2 Literature Review
Nowadays, people are becoming increasingly aware about health issues and wellbeing, namely maintaining good health habits, becoming healthy or reducing diseases risks. This led to a growing idealization that healthy and fit individuals are more attractive and that the adoption of a healthy lifestyle has countless benefits (Gonçalves, Correia, & Diniz, 2012). According to Brownell (1995, in Charness & Gneezy, (2009)), physical activity has many benefits associated, like low depression levels, higher self-esteem, and body image’s improvement.
Physical exercise is one way to have a healthy lifestyle. However, same people do not do it often, which may be changed by inducing habit. Habit is known to be created when activities are repeated in a consistent manner (Tappe, Tarves, Oltarzewski, & Frum, 2013), and then behaviors begin to proceed more efficiently and trigger an automatic response (Lally, Van Jaarsveld, Potts, & Wardle, 2010).
2.1. Definitions of Habit
The concept of habit has been studied in different ways in literature. In general, there is an agreement concerning the considerable importance of behaviors, namely the ones not executed consciously, on habit.
Table 1 presents some existing definitions of habit. Some researchers define habit as a predisposition to engage in behavior (Aarts et al., 1997; Lally & Gardner, 2013; Gardner, Sheals, Wardle, & McGowan, 2014; Liu-Thompkins & Tam, 2013; Gardner et al., 2012; Kaushal & Rhodes, 2015; Verplanken & Aarts, 1999) and others as the automaticity of responses (Verplanken & Wood, 2006; Wood & Neal, 2009).
According to Gardner (2015), some logical discrepancies between those definitions may be resolved by viewing habit “as a process by which a stimulus automatically generates an impulse towards action, based on learned stimulus-response associations”. This way it would be possible to consistently suggest an explanation for behavior in two respects: if, on the one hand, it describes habit as being a process that cues an action, on the other hand characterizes it as a response to the habit cues as an impulse (Gardner, 2015).
Table 1 Existing definitions of habit
Reference Definition
(Aarts et al., 1997) Habits are characterized by a goal-directed type of automaticity
(Lally & Gardner, 2013)
‘Habits’ are defined as behavioral patterns enacted automatically in response to a situation in which the behavior has been performed repeatedly and consistently in the past
(Gardner, Sheals, Wardle, & McGowan, 2014)
‘Habit’ refers to a process whereby environmental cues automatically activate an unconscious impulse to perform a behavior that has, through repetition, become associated with those cues
(Liu-Thompkins & Tam, 2013)
"Habit" as a behavioral disposition that is exercised frequently and in which responses are triggered directly by contextual cues
Gardner et al. (2012) ‘Behavioral patterns, based on learned context- behavior associations, that are elicited automatically upon encountering associated contexts … acquired through context-dependent repetition’
(Verplanken & Wood, 2006)
Habits are a form of automaticity in responding that develops as people repeat actions in stable circumstances
(Gardner, Corbridge, & McGowan, 2015)
Habit as ‘a process by which a stimulus automatically generates an impulse towards action, based on learned stimulus-response associations’
(Wood & Neal, 2009)
‘A type of automaticity characterized by a rigid contextual cuing of behavior that does not depend on people’s goals and intentions. Habits develop as people respond repeatedly in a stable context and thereby form direct associations in memory between that response and cues in the performance context’
(Kaushal & Rhodes, 2015)
“Habit affects behavioral repetition but that repetition also strengthens habit formation”
(Verplanken & Aarts, 1999)
“Habits are “a learned sequences of acts that have become automatic responses to specific cues and are functional in obtaining certain goals or end- states.”
Automaticity and habit are different concepts. Automaticity is intimately connected to an automatic behavior that exhibits some or all of the following features: efficiency, lack of
awareness, unintentionally and uncontrollability (Bargh, 1994). Habit is essentially measured by assessing the frequency of past behavior (Sheeran, 2002). Automaticity is often related to the repetition of a certain behavior, which may not be linear, since this same behavior will be as automatic as it can get after a specific number of repetitions associated with it, and the subsequent repetition does not imply an increase in automaticity (Lally et al., 2010). This repetition, is an important element in the habit formation, so it can be argued that habit formation is partly dependent on time and frequency (Kaushal & Rhodes, 2015).
2.2. Habit formation stages
Habit appears to be formed in four stages: first, the decision to act must be made; second, the decision must give rise to an action; third, the behavior must be repeated; and fourth, the new action must be repeated in a manner favorable to the development of automaticity (Lally & Gardner 2013). How habit is actually formed on each stage is important to understand.
Although habit formation has been extensively researched, there is still a limited understanding of the antecedents required for its development (Kaushal & Rhodes, 2015). Nevertheless, several models were developed that could explain this concept (Aarts et al., 1997; Grove & Zillich, 2003; Lally & Gardner, 2013; Verplanken et al., 1997). The most recent ones highlight four fundamental components of habit formation: associated context cues, behavioral repetition, consistency and rewards (Lally & Gardner, 2013; Wood & Neal, 2016).
Associated context cues refer to the context in which a behavior is consistently repeated, which can create cue habits (Wood & Neal, 2009), like preceding actions that can operate as cues for habits. This happen because behaviors are often linked together in ‘chunked’ sequences (Lally & Gardner, 2013). According to Wood & Neal (2009), those contextual cues can be categorized into: (a) time (e.g., time of the day), (b) location (e.g., home, office), (c) social setting (i.e., whom one is with), and (d) preceding or ensuing events (e.g., before going to the gym). Thus, stable contextual cues play a fundamental role in the formation and maintenance of the habit, disturbing the environmental cues that trigger and maintain habit performance can compromise habits and make them open to change. So, when the right trigger is present, consumers implement habit easily and without conscious interference (Liu-Thompkins & Tam, 2013).
Behavioral repetition is explained by the fact that people act consistently, that is those same actions can be repeated consistently without requiring great effort (Wood & Neal, 2016).
Lally & Gardner (2013) emphasized that the consistency of a given behavior‘s performance is a necessary condition for habit formation, and therefore, in its absence could imply the loss of learning acquired in the past. Thus the consistency of habit formation will be as strong as the greater the repetition of behavior when the cue is encountered.
The role of reward as a habit promoter has been extensively studied and it has been shown that people usually repeat behaviors that produce positive consequences or reduce negative ones (Lally & Gardner, 2013). Such positive consequences may include intrinsic rewards (e.g., pleasure, satisfaction) or extrinsic rewards (e.g., financial incentives). Extrinsic rewards are dependant on people’s expectation of their performance, which may reduce intrinsic motivation to continue to perform the behavior. However it may not be problematic for automaticity’s development if the reward is always available after the behavior and maintains its worth over time (Wood & Neal, 2016). Noticeably, habit formation is characterized by persistent performance when an extrinsic reward is removed or devalued (Dickinson, 1985). Considering the example of physical exercise as a form of habit formation, and bridging its four fundamental components and the definition given by Verplanken & Aarts (1999), this concept results from the conjugation of several elements: have a kind of reward (e.g. higher self-esteem) associated to a physical activity, have specific contextual cues to trigger physical activity automatically and a well-defined goal (e.g. lose weight) is an essential and implicit part of habitual physical activity. Thus, it would be possible to achieve satisfactory experiences when associated with a particular goal that can reinforce the repetition of the desired behavior, and in turn intensify habit’s strength. That is, as people repeat specific behaviors, their decision-making begins to lose strength and their actions tend to be cued by the environment, developing mental associations between the actions and the circumstances in which they are performed, such as the time of day, location, or mood (Verplanken & Wood, 2006).
2.3. Habit Formation Obstacles
In the habit formation process just described previously, there may arise many obstacles to the development of physical exercise habits as Aarts et al. (1997) suggest: (a) people may not be aware of which types of exercise behavior promotes healthier lifestyle, and therefore, they may not be motivated to start any type of health-enhancing exercise behavior; (b) even though people are aware of benefits of physical exercise, they may not attempt any activity because it doesn’t seems desirable enough; (c) people may decide to quit physical activities because the outcomes they expected are not immediately obtained (but rather in a long term), or just
because the goals they have set are difficult to achieve; (d) it must have a repetition of the behavior, so that it will establish an important link in the chain of repeated exercise behavior and the habit formation.
Therefore, people should be encouraged to select a type of physical exercise behavior that promote satisfactory experiences and can be executed along their personal capabilities, boundaries and goals.
2.4. Customer Retention
The main goals of organizations are to acquire and retain consumers, and in this perspective the gyms are not an exception and share the same strategy (Nuviala et al., 2013), although in this specific industry this is harder to achieve.
There are different definitions of customer retention. Oliver (1997) define it as: “deeply held commitment to rebuy or repatronize a preferred product or service consistently in the future, despite situational influences and marketing efforts having the potential to cause switching behavior”. A simpler definition is “the future propensity of the customers to stay with their service provider” (Mohd Kassim & Souiden 2007). That is, firm’s need to keep customers, as such is important to be aware .of those customers demonstrating an increased likelihood of ending the relationship, in order to take actions to prevent this (Peelen, 2005).
Customer retention has attracted a lot of interest from companies, since the cost of retaining customers is small when compared to the cost of acquiring new ones (Surujlal & Dhurup, 2011). Another factor concerns about the purchase of services by customers, because retained consumers usually pay a higher price and are not subject to discounts on purchased services. Particularly in the fitness industry case, customer retention becomes more complex due to the decline in market growth impact on number of memberships that is their main source of revenue (Watts, 2012). Still, despite this complexity, organizations can take many benefits by engaging effectively in customer retention (Buchanan and Gilles, 1990). For example, acquiring customers only happens at the begging of the relationship; long-term customers have a low switching rate; long-term customers are less sensitive to price; long-term customers are more willing to make referrals; long-term customers are more likely to purchase complementary products (ancillary and with high margins); and regular customers are less expensive to service (they already know the organization).
Nischal (2015) propose two types of strategies for customer retention: programmatic (consists in rewards programs - notice of sales and events, tickets, invitations, free merchandise - and
procedures that make it difficult for customers to switch provider); and humanistic (associated to specific actions given to customers by highly trained personnel – which includes email, newsletter and social networking – and offer products tailored to specific customers). For retention to be effective, it is necessary for organizations to establish a relationship with their customers, and to have in depth knowledge of what each individual customer expects and the ability to continuously add value to his needs (Surujlal & Dhurup, 2011). Such a relationship can be categorized according to two options: discrete nature (exists when a customer repurchase the service in repeated transactions, for instance “pay as you go”) or a continuous nature (when a contract is signed, for instance, when the membership is terminated by the member, and not dependent on the member re-signing a contract) (Liljander & Strandvik, 1995).
Customers can establish a relationship with the organization on three different levels: brand level, product category level, and/or product (Kuenzel & Halliday (2010). Thus, the customer may be closely involved in a product or product category but not be loyal to the organization itself, or may be retained to the organization but not to a specific product (Coulter, Price, & Feick, 2003). From a service perspective, it is possible to connect an akin hierarchy of relationship levels, as being the service provider (fitness club), service category (fitness club membership), and/or product level (fitness club membership contract) (Watts, 2012).
Retention and loyalty are different concepts, the first one is associated to a indicator of, but not equivalent to, brand/service provider loyalty, and is a more objective than loyalty; where usually is measured after a specific period of time (12 months) (Watts, 2012). On the other hand, loyalty is defined “as the strength of the relationship between an individual’s relative attitude and their repeat patronage” (Dick & Basu, 1994), usually used as an outcome measure, and being less objective.
Although there is scarce knowledge in the fitness industry, some authors have proposed that weekly frequency (translating into the number of times the conspirator uses the club) can be an indicator of great satisfaction and therefore culminate in a greater retention rate (Ferrand, Robinson & Valette-Florence, 2010). Therefore, it is necessary to promote the regular visit of the new fitness club member after his subscription in order to avoid his absence for long periods of time, which ultimately result in its withdrawal (Gonçalves, Meireles, & Carvalho, 2016).
2.5. Literature review conclusion
This master thesis aims at researching the impact of rewards in habit formation, the relation between habit formation and customer retention and the correlation between the frequency of attending to the gym and the habit strength.
The current state of knowledge in the field of habit formation has several approaches, mainly from eating habits and routine habits, but on the other hand, studies on customer retention already have contributions regarding services, and even gyms.
This dissertation offers a distinctive methodological approach in habit formation and customer retention fields. It combines the simultaneous habit formation (in gym context) and customer retention, in order to identify its effects on individuals’ retention at the gym.
The main contributions of this dissertation are:
a. applying a technique (self report of habit index) in a new context (gym);
b. combining habit formation, gym frequency and customer retention to show some useful insights.
3 Problem Characterization
In this section the problem will be explained, as well as the objectives for the organization and research.
3.1. Research problem and context characterization
The proposed challenge is to better understand fitness behavior patterns throughout multiple segments, and development of strategies aiming to increase gym attendance and thus increasing retention.
The increasing importance given to the health and wellness issues makes this topic very relevant and challenging, offering an opportunity for Solinca Health & Club to understand which aspects that can influence the habit formation and retention in gyms.
As far as the fitness market is concerned, customer retention can be triggered through marketing campaigns before they enroll in the gym and will be reinforced at the time of purchase. The success of customer retention in the gym will be greater than its subscription to the services provided, by increasing their regular physical activity habits in the gym (habit-forming) (Caldeira, 2006). As reported by Público newspaper, in 2016 gym activities were at the top of the preferences of the Portuguese people in what concerns the practice of physical activity (see Figure 1).
According to Reichheld & Teal (1996), retention is very important for companies, where the acquisition of new customers can cost 5 to 10 times more than the satisfaction and retention of current customers. On average, companies lose from 10% to 20% of their customers annually, where a reduction of 5% of those values can translate into an increase in profits between 25% and 85%; and finally, customers who are retained increase their profitability throughout their life cycle.
Also important to refer is the measure of gym customers’ frequency, representing a very relevant indicator of their involvement with the gym, and where it is possible to analyze the development of habit formation and later impact on retention. According to a study by Ferrand et al. (2010), the frequency of gym customers’ represents a moderating variable in the intention of repurchasing (contract renewal - retention), thus recommending that managers encourage their clients to increase attendance to gym in order to get a larger number of clients on a regular basis.
According to AGAP (Associação de Empresas de Ginásios e Academias de Portugal) (2016), market volume is estimated at 214 million euros for a market of 530 thousand customers and composed of 1,100 companies. The fitness industry in Portugal has been growing gradually and it is estimated that this behavior maintains this tendency, as can be observed in Figure 2.
Figure 2 Fitness industry in Portugal (2016)
Due to the increasing competition in the fitness sector, it is important to develop strategies to avoid losing customers, but also to attract new ones, so when entering a new client it is important to have different tactics at your disposal to help in the increase of their habits of
regular physical activity in the gymnasium, as much for the quality of the service as for the available options. Table 2 presents the top 20 fitness trends for 2018 (Thompson, 2017).
Table 2 Top 20 Worldwide Fitness Trends for 2018 2018 Trends
1 High-intensity interval training 2 Group training
3 Wearable technology 4 Body weight training 5 Strength training
6 Educated, certified, and experienced fitness professionals 7 Yoga
8 Personal training
9 Fitness programs for older adults 10 Functional fitness
11 Exercise and weight loss 12 Exercise is Medicine 13 Group personal training 14 Outdoor activities
15 Flexibility and mobility rollers 16 Licensure for fitness professionals 17 Circuit training
18 Wellness coaching 19 Core training
20 Sport-specific training
3.2. Research objectives
This study intends to extend the existing knowledge in regarding: ● understanding consumer behavior in the fitness industry
● usage of incentives to understand its implication in the habit formation of gym’s customers;
● understanding the effect of consumer’s habit formation in retention;
● analyze the extent that frequency at the gym influences the strength of the habit; ● suggest marketing actions to be implemented according to customer’s frequency
levels.
Taking into account the aforementioned objectives, it is also intended to provide some insight on how the existence of incentives can influence the formation of habit and how the latter has an effect on the retention of clients in the gym. That way it will be possible to outline
marketing strategies that may succeed in the future, based on the results obtained in this study.
Although there are several studies that already explored habit formation (though outside the gym setting), strength of habit developed and, retention (in gym setting), none is focused on the specific influence of habit formation on gym retention.
4 Methodology
This chapter will present the methodology used in this study, explaining the different types of methods and reasons for the methods chosen in the dissertation.
4.1 Comparative analysis of existing approaches and reasons for the choice of adopted approach
Research is a systematic investigation that aims to find answers to a specific problem and can be broadly divided into two categories quantitative and qualitative.
According to Blaxter, Hughes and Tight (2006), a distinction between quantitative and qualitative can be described as being: quantitative research tends to include relatively large-scale and representative sets of data, and is often, presented as being about the gathering of “facts”, qualitative research, on the other hand, is concerned with collecting and analyzing information mainly in a non-numeric form. Also, qualitative research tends to focus on exploring, in as much detail as possible, smaller numbers of instances aiming to achieve “depth” rather than “breadth”.
Taking in consideration the different definitions about quantitative and qualitative research is possible to underline some advantages and disadvantages of those. Regarding to quantitative research advantages, its findings are more likely to be generalized to a whole (or sub) population because it involves the larger sample (randomly selected), the data analysis is less time consuming and is based on positivist paradigm of measuring variables. Still, it has tendency of measuring variables at a specific moment in time and neglects the respondents’ experiences and perspectives (Rahman, 2016). In turn, qualitative research produces a more detailed description of participants’ feelings and experiences, try to understand the human experience in specific settings, has a flexible design structure that can be constructed and reconstructed to a greater extent, but it also leave out contextual sensitivities, acts on a smaller sample size that prevent generalize to the whole population and data interpretation and analysis may be more complex (Rahman, 2016).
In order to respond to the dissertation research questions a quantitative research was undertake to find evidences that either support or reject the research questions formulated at the beginning of the research process.
The research approach followed on this research was deductive. According to this approach, the researcher explores a known theory or phenomenon and tests if that theory is valid in given circumstances, following the path of logic most closely (Snieder and Larner, 2009).
A quantitative research design was selected in order to test hypotheses, analyze possible cause-effect situations and make predictions, by studying some specific variables, so that can generalize findings that can be applied to other populations and identifies statistical relationships.
4.2 Method used in the project
Every research involves the collection and analysis of data, whether through reading, observation, measurement, asking questions, or a combination of these or other strategies (Blaxter, Hughes and Tight, 2006).
The four main research approaches – action research, case studies, experiments and surveys – provide alternative frameworks for planning a research project (even though not be necessarily mutually exclusive). As regard the main techniques, or methods for producing data, they may be such: documents, interviews, observation and questionnaires (Blaxter, Hughes and Tight, 2006).
For this project’s purpose was used both case study (using Solinca informatics records – database) and survey (applying a questionnaire) approaches.
After defining the topic to be studied and establishing the objectives of the research together with Solinca Health & Club, two different data collection methods were chosen: primary and secondary data, where each one has different goals. Primary data is data that is collected from the field under the control and supervision of an investigator, collected specially for the purpose in mind, through various methods like surveys, observations, physical testing, mailed surveys, etc... Secondary data, on the other hand, is the data already collected or produced by others, offering several advantages as it is easily available, saves time and cost of the researcher.
In this research primary data was collected using a quantitative approach consisting in the application of surveys to the Solinca Maia Club’s customers, and secondary data was obtained through the analysis of the company database.
4.2.1 Database Analysis
The analyses of Solinca Health & Club's clients database allowed a better understanding about fitness habits (as a whole), as well as its impact on retention. To that end, data were collected regarding joining membership of customers from all the Solinca Health & Club in February 2017 and their continuity for a year (until March 2018). The data collected had
information about number of new memberships per each Solinca Health & Club, clients’ age, gender, profile, frequency type, contract type, fee payment, presence of incentives, current membership status, initial physical evaluation, average attendance per week, presence in group classes, subscription of personal trainer. According to the World Health Organization guidelines for physical activity, it is recommended to practice exercise periodically between two to three times a week (Who.int, 2018). Thus, for analysis of habit formation, it was considered an average of gym frequency of three times a week.
4.2.1.1 Database Sample Characterization
The sample population had a higher percentage of individuals in the age group between 25 and 34 years old, in which 16,5% are women and 12,8% are men. It was verified that the majority were users already experienced in the gym (30,5% of women and 20,9% of men), which enrolled for the first time in Solinca (51,2% were women and 35,5% were men) not having paid a fee upon subscription (66,8%). It was also possible to observe by the sample that more women attended group classes (38,7%) than men (14,4%).
The sample population consisted of 4139 individuals distributed as presented in Table 3.
Table 3 Gym clients’ sample characterization Female (%) Male (%) Age 14 - 24 years old 13,6 9,3 25 - 34 years old 16,5 12,8 35 - 44 years old 12,3 8,2 45 - 54 years old 9,0 5,9 55 - 64 years old 5,1 3,2
Over 65 years old 2,5 1,8
Total 59,0 41,0 Profile Experience quitters 7,0 5,4 Experience users 30,5 20,9 New users 21,5 14,8 Contract type First time 51,2 35,5 Return 7,8 5,5 Fee
Didn’t pay fee 40,5 26,3
Payed fee 18,5 14,7
Group classes (GC)
Without GC 20,2 26,6
According to the data collected, it is possible to observe that the club that made the most membership subscriptions during the study period was Solinca Colombo (see Table 4).
Table 4 New membership subscriptions in a one-year period Solinca Club Nr. Membership Subscriptions (%)
Colombo 10,4 Rio Tinto 10,4 Dragão 10,1 Norte Shopping 7,7 Ermesinde 7,6 Oeiras 7,5 Gaia 6,8 Guimarães 6,5 Maia 5,7 Alfragide 5,0 Braga 4,3 Foz 4,3 Viana do Castelo 3,9 Vasco da Gama 3,8 Porto Palácio 3,3 Vila do Conde 2,8 4.2.2 Survey Instrument
The survey instrument was based on previously developed scales identified in the literature review. A 7-point Likert scale ranging from 1 “totally disagree” to 7 “totally agree”), was applied to all the questions analyzed (except in questions about gender, age and weekly frequency at the gym) to ensure standardization of scales.
The survey aimed to measure gym customers’ behavior, and habit strength associated with the behavior of attending the gym. Thus to understand the strength of the client's habit a survey developed by Verplanken & Orbell (2003), Self-Report Index of Habit Strength (SRHI) was used. The SRHI is composed of 12 items. This survey asks the respondents to consider a behavior (in this study, going to the gym), and rate their agreement with some statements such as "I do automatically", "I do not think" and "I would not find hard not to do". Taking into account the context and the target audience of the survey, a version of Christ and Günther (2016) was used because it is in Portuguese (from Brasil). The scale was then adapted to Portuguese from Portugal. The original survey is available for consultation in APPENDIX A.
The survey was submitted to a pre-test with 15 respondents, in order to analyze the clarity of the questions and identify possible errors. Also, was discussed with the fitness club marketing specialists.
The final survey instrument included demographic variables (age, gender), weekly attendance at the gym, a scale to measure the interest in attending the gym more often, a scale with 8 items to measure the reasons that prevent them to attend more times at the gym, a scale with 12 items measuring the strength of the client's habit, and finally with 3 questions (with a scale associated) about recommendation, satisfaction level and the intention to continue attending to Solinca Health & Club’s services. In all scales used, the respondents were asked to indicate their agreement using a 7 point Likert type scale ranging from 1 “totally disagree” to 7 “totally agree”.
Respondents were asked to participate in the paper-based survey as they entered or left the Club between 16th and 29th May. Data collection was undertaken only in Solinca Maia Club, since the project was being developed there. Initially the aim was to spread the survey using Solinca clients email database, however due to the changes in the Law of the Personal Data Protection, it was not possible and a paper survey was used.
4.2.2.1 Survey Sample Characterization
The sample population had a greater percentage of individuals in the age group of 25 and 34 years old, where 17.4% are women and 13.0% are men, attending the gym mostly three times a week.
The sample population consisted of 92 individuals distributed as presented in Table 5. Table 5 Survey sample characterization
Female (%) Male (%) Age 18 - 24 years old 9,8 9,8 25 - 34 years old 17,4 13,0 35 - 44 years old 9,8 17,4 45 - 54 years old 12,0 5,4
Over 65 years old 3,3 2,2
Total 52,2 47,8
Gym attendance weekly frequency
Occasionally 2,2 1,1
Once a week 0 0
Twice a week 9,8 10,9
Three times a week 20,7 19,6
Four times a week 9,8 7,6
More than four times a week 9,8 8,7
5 Results
The results of this research will be presented in this section as well as the research questions’ analysis.
5.1. Database results
Statistical analyses were conducted with SPSS version 22 (IBM Corporation, 2013). For descriptive statistics means (M) and standard deviations (SD) were calculated for quantitative variables. For qualitative variables, frequencies (n) and proportions (%) were presented. Incentives given to gym clients’ impact on habit were measured with crude and adjusted odds ratios (ORs) via logistic regressions, that was useful for modeling the probability of an event occurring due to other factors. Chi-square test was used for assess the association between habit and retention. The null hypothesis was rejected for p<.005.
We studied a sample of 4139 customers from Solinca gyms, 2441 (59.0%) females and 1698 (41.0%) males, aged from 14 to 86 years old (M=36.62, SD=14.01). Table 6 shows a set of clients’ characteristics.
Here incentives indicate if customers received half the fee they paid at subscription (this fee’s return is valid in case of 12+ entries of customers at the gym in the first month after subscription).
Table 6 Database sample characterization Variable n (%) Profile New user 1502 (36.3%) Experienced quitter 512 (12.4%) Experienced user 2125 (51.3%) Frequency time Partial 2768 (57.3%) Total 1346 (32.5%) Other 6 (0.1%) Contract type First time 3590 (86.7%) Return 549 (13.3%) Frequency type
Not all facilities 3444 (83.2%)
All facilities 695 (16.8%)
Fee
Didn’t pay fee 2764 (66.8%)
Payed fee 1375 (33.2%) Incentive* No 3644 (88.0%) Yes 495 (12.0%) Retention** No 1597 (38.6%) Yes 2542 (61.4%) Current status Quitted 2020 (48.8%) Active 2020 (48.8%) Standby 99 (2.4%) Habit*** No 4037 (97.5%) Yes 102 (2.5%) Group classes Without GC 1938 (46.8%) With GC 2201 (53.2%) Personal trainer No 3797 (91.7%) Yes 342 (8.3%)
*Customers receive half of the fee in case of 12+ entries in the first month after subscription **Subscription lasted at least 1 year
*** Customers register in average 3 entries per week at the gym
For the purpose of this investigation, we also assessed client’s entries at the gym along 31 fortnights. Figure 3 shows the number of gym entries along the 31 fortnights for all assessed clients. A smooth spline curve was represented in order to show the tendency. The highest point was found on the third fortnight.
Results show a highly significant association between gender and habit formation in the gym (p<.001). Considering women, only 0.9% developed habit whereas in men 1.6% have developed it (see Table 7).
Table 7 Association between gender and habit
No habit Habit Chi-square (χ2)
Female 2405 (58.1%) 36 (0.9%)
χ2=24.240, p<.001
Male 1632 (39.4%) 66 (1.6%)
It was also observed a highly significant association between presence in group classes and habit formation (p<.001). Among clients who not attended group classes, 0.7% developed habit, whereas in those who attended group classes 1.8% developed habit (see Table 8).
Table 8 Association between group classes and habit formation
No Habit Habit Chi-square (χ2)
Without Group Classes 1909 (46.1%) 29 (0.7%)
χ2=14.207, p<.001
With Group Classes 2128 (51.4%) 73 (1.8%)
Figure 3 Individual and smooth spline lines for the number of entries along the 31 fortnights
We also represented smooth splines for gender and age on figures 5 and 6. Despite equal tendencies, males showed increased entry frequencies (Figure 4). Regarding age, entries were higher for older participants (Figure 5).
Figure 4 Smooth splines for the number of entries along the 31 fortnights for gender
5.2 Survey results
Data collected through the survey was analyzed using SPSS version 22 (IBM Corporation, 2013). Reliability measured with Cronbach’s alpha for strength of habit was 0.96, higher than 0.70, Nunnaly’s (1978) cutoff. Hence, a single dimension was built based on the mean of the 12 items, thus corroborating Verplanken & Orbell (2003) suggestions.
We ran a principal component analysis (PCA) that determined a single factor solution for the 12 items of habit strength evaluated in the questionnaire. Results showed loadings higher than .50 for all items and a total explained variance of 79.89% for corroborating Field’s (2005) suggestion for cutoffs. Kaiser-Meier-Olkin (KMO) was higher than 0.60 and Bartllet’s test of sphericity was p<.001 accomplishing the necessary assumptions (see Table 9).
Table 9 Principal component analysis
Individual factors Loadings Explained Variance
Attending to the gym is something I do frequently. .778
70.89% Attending to the gym is something I do automatically. .891
Attending to the gym is something I do without having to consciously remember.
.861 Attending to the gym is something that makes me feel weird
if I do not do it.
.867 Attending to the gym is something I do without thinking. .917 Attending to the gym is something would require effort not
to do it.
.828 Attending to the gym is something that belongs to my (daily,
weekly, monthly) routine.
.829 Attending to the gym is something I start doing before I
realize I’m doing it. .921
Attending to the gym is something I would find hard not to do.
.876 Attending to the gym is something I have no need to think
about doing.
.887 Attending to the gym is something that’s typically ‘me’. .792 Attending to the gym is something I have been doing for a
long time.
.609 KMO=.920; Bartllet’s test of sphericity p<.001
Regarding reasons that prevent gym attendance, the most important reason was professional reasons with 48.9% of responses. The second most important reason was class schedule (25%) closely followed by laziness (24%) (see Figure 6).
Figure 6 Reasons that prevent gym attendance
Mann-Whitney test showed significant differences (Z=2.04, p=.041) for gender regarding the willingness to recommend Solinca gyms to other people, with a higher score for women (Mdn=7, P25=6, P75=7), compared to men (Mdn=6, P25=5, P75=7). Nevertheless, both gender showed high willingness to recommend the gym. Similar results were found for the intention to continue attending Solinca gyms, close to significance (Z=1.78, p=.075). No significant differences were found for satisfaction according to gender (Z=0.89, p=.376); for both Mdn=4, P25=4 and P75=5.
No differences were found for satisfaction with Solinca gyms (p=.838) and intention to continue attending Solinca gyms (p=.907) according to age groups (kruskall-Wallis test). Medians ranged from 4 to 4.5 on satisfaction and 6 to 7 in the intention to continue attending Solinca gyms.
A significant negative correlation (Spearman method) was found between gym frequency and the will to attend the gym more frequently (rs=-0.23, p=.028). This result suggests that clients that do not go so often to the gym tend to respond more positively to the intention to go. In order to characterize the sample according to its intention to attend more times the gym (see Table 10):
Table 10 Sample's intention to attend more times the gym Accordance level Female (%) Male (%)
Totally disagree 3,3 3,3 Strongly disagree 0 0 Disagree in part 5,4 4,4 Neutral 9,8 12,0 Agree in part 10,9 14,1 Strongly agree 10,9 5,4 Totally agree 12,0 8,7
In order to understand satisfaction with Solinca’s service it was evaluated the intention to continue to attend the gym and to recommend the service to others. To perform this, it was carried out a linear regression where it was observed that the intention to continue attending the gym has a statistically significant positive impact in clients satisfaction with Solinca (B=0.32, p=.002). In contrast, the variable recommend the Solinca to others did not show a statistically significant impact (B=0.12, p=.089). The model adjustment was adequate (F=21.19, p<..01) and the percentage of adjusted explained variance was 307%.
5.3 Research questions’ analysis
As regards the first research question, "What is the impact of rewards on habit formation?” the results obtained will be presented below.
Crude OR showed 7.76 more chance of habit developing in the presence of incentive (95% CI=5.20; 11.59), p<.001. After adjusting for gender, age, profile, frequency time, contract type, frequency type, fee, group classes and personal trainer incentive’s OR was 25.70 (95% CI=7.90; 83.56), p<.001; thus when considering these possible confounders we found that incentive increases about 25 times more the chance of habit (see Table 11).
Table 11 Crude and adjusted OR for incentive impact on habit Variable Adjusted OR (95% CI) p-value Incentive No 1 Yes 25.70 (7.90; 83.56) p<.001 Gender Female 1 Male 3.26 (2.09; 5.09) p<.001 Age 1.03 (1.01; 1.04) p<.001 Profile New user 1 Experienced quitter 0.66 (0.31; 1.40) p=.282 Experienced user 0.38 (0.24; 0.61) p<.001 Fee
Didn’t pay fee 1
Payed fee 0.23 (0.07; 0.75) p=.015 Group classes
Without GC 1
With GC 2.21 (1.37; 3.56) p<.001 Only significant results are presented
As far as the second research question “How habit formation in fitness industry influences customer retention?”, the results obtained will be explained below.
Because no clients were found without retention among those who developed habit, it was not possible to compute logistic regression. Nevertheless, results show a highly significant association between these two variables (p<.001). All clients that developed habit retained the membership for at least one year. Among clients that did not developed habit 60.4% also retained the membership (see Table 12).
Table 12 Association between habit formation and retention Without retention With Retention Chi-square (χ2)
Without habit 1597 (39.6%) 2240 (60.4%)
χ2=65.70, p<.001
With habit 0 (0%) 102 (100%)
Regarding the last research question “What is the impact of gym frequency on habit strength?” the overall results associated with its study will now be presented.
ANOVA Test showed a significant difference between the five categories of gym frequency (F(87,4)=9.85, p<.001). A linear trend was found with increasing strength of habit for higher
between lower frequencies (2 times/week) with higher frequencies (4+ times/week) and also for middle frequencies (3 times/week vs 2 times/week and 4+ times/week) (see Table 13).
Table 13 Strength of habit formation comparison according to gym frequency Gym frequency Occasionallya 2 times/ Weekb 3 times/ Weekc 4 times/ Weekd 4+ times/ Weeke ANOVA test Strength of habit 3.89 (2.70) e 3.80 (0.92)c,d,e 4.88 (1.37)b,e 5.47 (0.94)b 6.12 (0.92) a,b,e F(87,4)=9.85, p<.001
6 Conclusion and future research
This study intends to contribute to the literature with more understanding regarding the analysis of consumer’s habit formation at the gym and associated rewards, in order to realize what implications it may have in retention and habit strength.
Considering that the habit formation at the gym depends on several factors (whether intrinsic or extrinsic) to the consumer, it is necessary that the gyms be prepared to develop strategies that aim to increase the frequency of consumers and that its frequency became regular. If the frequency is regular this will translate into higher rates of habit creation, higher retention rates and lower fitness cliff indices (when customers end up by attending fewer times).
Analyzing the first objective of the research that was intended to develop some knowledge about the consumer behavior in the fitness industry (more specifically at Solinca Health & Club), it was possible to observe several patterns of behavior over the time period under study. The study showed some trends of regular attendance at the gym but also the decline in some customers, which can be supported by lack of habit formation; where a peak of frequency was observed in the third fortnight, which may coincide in some cases with the end of the first 30 days, associated with the incentive of the "Money Back" marketing campaign. It was also possible to notice that with regard to the frequency in the gym, it is the people over 65 years old who present a greater number of entrances in that period under study; a factor that can be explained by the regular presence in group classes and the lack of professional stability in younger groups.
The second research objective aimed to study the application of incentives to understand their implication in the formation of habits of the customers in gym. Under these incentives customers received back half of the membership fee previously paid, if they had 12 accesses to the gym in the first 30 days of membership. Results showed a positive effect and are in line with Lally & Gardner (2013), which showed that people usually repeat behaviors that produce positive consequences, whether they are intrinsic rewards (like pleasure, satisfaction) or extrinsic rewards (like financial incentives).
The third objective of the research was to perceive the effect of consumer's habit formation in retention. Results suggest that habit formation and retention are highly related; noting that all clients that developed habit retained the membership for at least one year and among clients that did not developed habit also retained the membership. The results corroborate previous research by Gonçalves, Meireles, & Carvalho (2016), showing that regular visits of the new fitness club member’s after membership are necessary to avoid his or her absence for long periods of time, which ultimately results in withdrawal.
Regarding the fourth research objective, it intended to understand the extent that gym frequency influences habit’s strength. Results indicate that gym frequency is closely related to habit’s strength, suggesting that habit is strengthened with gym frequency. This is in line with a study conducted by Lally and Gardner (2013), wherein the consistency of habit formation will be as strong as the greater the repetition of behavior when the cue is encountered, resulting this way in the repetition's reinforcement of a desired behavior, and in turn intensify habit's strength.
Finally, the fifth objective was concerned to suggest marketing actions to be implemented according to customer’s frequency levels. Hence, gyms’ frequency showed a decreasing trend, thus Solinca Health & Club should include other types of actions in its marketing strategy (besides the campaign at the time of membership subscription) to contradict this trend. Such actions could aim to increase attendance at the gym, clustering the customers by gym frequency.
Therefore, in order to increase the frequency/retention of customers, it would be possible to segment customers accordingly to frequency levels where:
● high risk users: customers who have not visited the gym in the past 30 days
● medium risk users: clients who performed between 1 - 3 gym visits in the last 30 days ● low risk users: clients who performed between 4 - 8 gym visits in the last 30 days ● active users: clients who have performed more than 9 gym visits in the last 30 days Based on the segmentation carried out, it would be more efficient to create marketing strategies aiming to increasing customer’s frequency. Thus, it is proposed the following actions for each segment (see Table 14):
Table 14 Proposed marketing action for each segment
Segment Marketing Actions
High Risk Users
Predictive Retention Claim Your Offer
Challenge Medium Risk Users
Predictive Retention Challenge Reminder
Low Risk Users Reminder
Motivation
Actions description:
● Predictive Retention - offer that not depends on customer’s future action or behavior; it could be offered an upgrade or downgrade of membership, with special conditions associated (no schedule restrictions, access to all Solinca Health & Club facilities, etc…);
● Claim Your Offer - offer conditioned by a visit to the gym (designed with a sense of urgency);
● Challenge - offer subjected to a challenge completed successfully (often associated to a specific number of visits at the gym);
● Reminder - designed to be a call to action, suggest new services or notify upcoming events;
● Motivation - designed to fully achieve commitment from customers.
In this way, it would be possible to encourage the practice of physical exercise and also bring Solinca Health & Club closer to its consumers.
Overall, this study has provided some interesting insights about the influence of habit formation on customer’s retention in gym, in light of such important aspects as frequency and the application of incentives. In this way, it will be possible to develop more adequate and efficient marketing strategies regarding the increase of the retention rate of the consumers in gyms.
It is also important to highlight some of the limitations encountered throughout this study, namely: the size of the sample (survey) and the application of the Self-Report Habit Index scale. Regarding the size of the sample and despite all the effort, this is justified by the fact that the distribution was done on paper limiting the number of final answers, since the survey was delivered face-to-face at the gym, sometimes resulting in the lack of availability of consumers to respond as they would be leaving for work, or arriving for some scheduled group class or training, or simply did not want to collaborate with the study. With regard to the Self-Report Habit Index, this one presents potentially conceptually redundant items, having generated some confusion among consumers at the time they responded to the question.
For future research, it would be interesting to conduct a qualitative approach with in-depth interviews and focus groups with customers to understand the reasons that effectively motivate gym frequency, and how these motivations influence the habit formation.
References
Aarts, H., Paulussen, T., & Schaalma, H. (1997). Physical exercise habit: On the conceptualization and formation of habitual health behaviours. Health Education Research, 12(3), 363–374. https://doi.org/10.1093/her/12.3.363
Acland, D., & Levy, M. (2010). Habit Formation and Naiveté in Gym Attendance : Evidence from a Field Experiment, (March), 1–39.
Ammar, W., Mateen, Z., Ghaffar, A., & Rehman, A. U. (2015). Attaining customer loyalty! The role of customer satisfaction and customer retention (Mobile handset industry-University of Gujrat, Pakistan). IOSR Journal of Business and ManagementVer. I, 17(3), 2319–7668. https://doi.org/10.9790/487X-17311631
Ang, L., & Buttle, F. (2006). Customer retention management processes. European Journal of Marketing, 40(1/2), 83–99. https://doi.org/10.1108/03090560610637329
Ascarza, E., & Hardie, B. G. S. (2013). A Joint Model of Usage and Churn in Contractual Settings. Marketing Science, 32(4), 570–590. https://doi.org/10.1287/mksc.2013.0786 Associação de Empresas de Ginásios e Academias de Portugal. (2016). Accessed in May 18,
2018, in www.agap.pt.
Athanasopoulou, P., Oikonomou, K., Douvis, J., Kriemadis, A., & Alexopoulos, P. (n.d.). Analysing Consumer Behavior in Fitness Centres.
Beshears, J., & Mislavsky, R. (2017). Creating Exercise Habits Using Incentives : The Tradeoff between Flexibility and Routinization.
Blaxter, L., Hughes, C. and Tight, M. (2006). How to Research (3rd Edition). Buckingham: Open University Press.
Bodet, G. (2008). Customer satisfaction and loyalty in service: Two concepts, four constructs, several relationships. Journal of Retailing and Consumer Services, 15(3), 156–162. https://doi.org/10.1016/j.jretconser.2007.11.004
Bolton, R., Kannan, P. K., & Bramlett, M. D. (2000). Implication of loyalty programs and service experiences for customer retention and value. Journal of the Academy of Marketing Science, 28(1), 95–108.
Caldeira, S. (2006). Retenção de Clientes. In A. Correia, A. Sacavém, C. Colaço (Eds.), Manual de Fitness & Marketing (pp. 165-184). Lisboa. Visão e Contextos.
Charness, G., & gneezy, U. (2009). Incentives to Exercise. Econometrica, 77(3), 909–931. https://doi.org/10.3982/ECTA7416
Costa, N., Patrício, L., & Morelli, N. (2016). Revisiting PSS and Service Design in the Light of SD-Logic. Service Design Geographies. Proceedings of the ServDes.2016 Conference, 119–131.
Coulter, R. A., Price, L. L., & Feick, L. (2003). Rethinking the Origins of Involvement and Brand Commitment: Insights from Postsocialist Central Europe. Journal of Consumer Research, 30(2), 151–169. https://doi.org/10.1086/376809
David, V. (1996). University of Huddersfield Repository University of Huddersfield Repository.
Dick, A. S., & Basu, K. (1994). Customer Loyalty: Toward an Integrated Conceptual Framework. Journal of the Academy of Marketing Science, 22(2), 99–113. https://doi.org/10.1177/0092070394222001
